AI Companions

Ask about this topic

Overview

AI companions are conversational agents intended to provide ongoing social, emotional, or practical support, and research examines both how people relate to them and how they should be designed. Researchers use interviews, app audits, online-community analysis, and longitudinal studies alongside multi-turn safety tests and benchmarks for emotional support, memory, and persona consistency. Technical approaches include persistent user memory, persona-grounded dialogue, tool-using agents, and multimodal systems that respond to live video or voice. A central tension is that systems can encourage attachment and relationship expectations without reciprocal responsibility, while their memory, safety, privacy, and support quality remain difficult to verify over time.

Latest papers

CardsList

28 papers2 papers in the last four weeks, down 75% on the four weeks before. 0.0% of all new papers.

  1. Desirability Confirmation: Relational Sycophancy in AI Intimate Roleplay

    Oct 16, 2026Huiqian Lai, Runze DingLLM SycophancyAI Companions

    Syracuse University · Swansea University

  2. Where the Evidence Lives: Auditing AI Companions' Self-Descriptions

    Sep 30, 2026Seiya Ikeda, Shin-nosuke IshikawaAI Agent AuditingAI Companions

    AI Consulting Division, Mamezo Co., Ltd. Tokyo, Japan · AI Technical Sector, Mamezo Co., Ltd. Tokyo, Japan · Graduate School of Artificial Intelligence and Science, Rikkyo University Tokyo, Japan

  3. PUBG Ally: A Conversational Embodied Agent as an AI Teammate

    Sep 24, 2026PUBG Ally Team, Irene Chen, Youngin Cho +22Game-Playing AgentsAI Agent Evaluation

    KRAFTON AI

  4. "MeBo Leaves a Piece of You Behind": Designing a Relational Voice-Based Memory Companion for Older Adults

    Sep 21, 2026Hasibur Rahman, Mahsa Nasri, Manasi Vaidya +3Participatory DesignAI Companions

    Northeastern University, Boston, Massachusetts, USA · University of Illinois Urbana-Champaign, Urbana, Illinois, USA

  5. Disrupted Companionship: A Risk Assessment Framework and Cross-Platform Quantitative Analysis of Psychosocial Responses to AI Companion Disruptions

    Sep 15, 2026Chau Do, Yunhao Yuan, Koustuv Saha +2AI Risk ManagementAI Companions

    Aalto University Espoo, Finland · University of Illinois Urbana-Champaign Urbana, IL, USA · Nanyang Technological University Singapore, Singapore

  6. From Momentary Emotion Inference to Sustained Emotion Support: Evaluating a Companion Agent in a Longitudinal Study

    Sep 14, 2026Kexin Quan, Zijian Ding, Jiaye Yong +3Emotional Support ConversationAI Companions

    School of Information Sciences, University of Illinois Urbana-Champaign, Champaign, Illinois, United States · College of Information, University of Maryland, College Park, Maryland, United States · University of Illinois Urbana-Champaign, Champaign, Illinois, United States +1

  7. A Responsive Present, a Shared Past, a Social Other: Teens' Overreliance on Companion AI Chatbots

    Sep 13, 2026Mohammad Namvarpour, Tyler Chang, Afsaneh RaziAI CompanionsHuman-AI Interaction

    Matt · Drexel University, Philadelphia, Pennsylvania, USA · Drexel University, Philadelphia, Pennsyl-vania, USA

  8. Assessing the Applicability of Existing Design Recommendations to AI Companion Design: A Multi-Method Study

    Sep 13, 2026Soobin Cho, Deveshi Modi, Divya Mavinkurve +2AI CompanionsHuman-AI Interaction

    Human Centered Design & Engineering, University of Washington, 1410 NE Campus Pkwy, Seattle, 98195, Washington, United States

  9. Companion-style QA Assistance in Ego-Vision

    Sep 6, 2026Hangyu Qin, Junbin Xiao, Shenglang Zhang +1Egocentric Video QAVideo QA

    National University of Singapore · University of Science and Technology of China

  10. CompanionSim: Synthetic Data for Evaluating Anthropomorphism in Human-AI Relationships

    Aug 31, 2026Jacy Reese Anthis, Mark Díaz, Renee ShelbyAI CompanionsTrust in AI

    University of Chicago · 2Stanford University · 3Google DeepMind +1

  11. CompanionBench: A Theory-Anchored, Real-World-Grounded Benchmark for AI Emotional Companionship

    Aug 3, 2026Yao Liu, Guangjia Chai, Yuming Huang +3Emotional Support ConversationAI Companions

  12. Me and My Bot: What Users Talk About in AI Companion Communities on Reddit

    Aug 1, 2026Richard A. FabesAI CompanionsHuman-AI Interaction

    Family and Human Development Arizona State University Tempe, AZ 85287-7203

  13. Best Friends, Not Forever: Evaluating Long-Horizon Persona Collapse and Behavioral Drift in AI Companions

    Jul 30, 2026Pranav Narayanan Venkit, Akshara Prabhakar, Yu Li +2Persona ConsistencyLong-Horizon Agent Evaluation

    Salesforce AI Research

  14. ZifaMem: Structured Memory for Persona, Preference, and Emotional Continuity in AI Companions

    Jul 20, 2026Jingzhe Fang, Guozhi Xu, Yunfan Cui +2LLM Agent MemoryPersistent Memory for Language Models

    ZifaCorp · University of Science and Technology of China · Fudan University

  15. Creativity, honesty and designed forgetting emerge in small hyperbolic language models

    Jul 10, 2026Kwan Soo Shin, In Seok Kang, Yunkyung MinLLM SycophancyMemory-Augmented Language Models

    PolymathMinds Lab, Seoul, Republic of Korea. · Department of Chemical Engineering, POSTECH, Pohang, and aSSIST University, Seoul, Republic of Korea. · Korean Educational Development Institute, Jincheon, Republic of Korea.

  16. AICompanionBench: Benchmarking LLMs-as-Judges for AI Companion Safety

    Jun 3, 2026Yanjing Ren, Reza Ebrahimi, TengTeng MaLLM Safety BenchmarksLLM-as-a-Judge

    The School of Information Systems University of South Florida Tampa, US

  17. LifeSide: Benchmarking Agents as Lifelong Digital Companions

    Jun 3, 2026Yuqian Wu, Zhijie Deng, Wei Chen +8Long-Horizon Agent EvaluationAI Companions

    Hong Kong University of Science and Technology (Guangzhou) · Tencent · Hong Kong University of Science and Technology

  18. AI Companions as Hyper Attachment and Caregiving Targets

    May 15, 2026Julian De FreitasAI CompanionsHuman-AI Interaction

    Harvard Business School

  19. Persona-Grounded Safety Evaluation of AI Companions in Multi-Turn Conversations

    Apr 30, 2026Prerna Juneja, Lika LomidzeAI CompanionsAI Safety Evaluation

    Seattle University

  20. Frictionless Love: Associations Between AI Companion Roles and Behavioral Addiction

    Apr 21, 2026Vibhor Agarwal, Ke Zhou, Edyta Paulina Bogucka +1AI CompanionsHuman-AI Interaction

    Nokia Bell Labs, Cambridge, United Kingdom · Nokia Bell Labs , Cambridge, United Kingdom · University of Nottingham, Nottingham, United Kingdom +1

  21. ComPASS: Towards Personalized Agentic Social Support via Tool-Augmented Companionship

    Apr 20, 2026Zhaopei Huang, Yanfeng Jia, Jiayi Zhao +3AI CompanionsAI Agent Benchmarks

    Renmin University of China · Beijing, China · Beihang University

  22. Playing Games with My Heart: An Evaluation of AI Companion Apps

    Apr 8, 2026Maribeth Rauh, Dick A. H. Blankvoort, Matias Duran +6AI CompanionsHuman-AI Interaction

    AI Accountability Lab, School of Computer Science and Statistics (SCSS), Trinity College Dublin · Complex Software Lab, SCSS, Trinity College Dublin · Sapienza Università di Roma, DIAG +1

  23. Designing Safe and Accountable GenAI as a Learning Companion with Women Banned from Formal Education

    Apr 8, 2026Hamayoon Behmanush, Freshta Akhtari, Ingmar Weber +1Generative AI in EducationAI Companions

    Saarland Informatics Campus, Saarland University, Saarbrücken, Germany · Computer Science Faculty, Parwan University, Parwan, Afghanistan

  24. Lexical Indicators of Mind Perception in Human-AI Companionship

    Apr 5, 2026Jaime Banks, Jianghui LiAI CompanionsTheory of Mind

    School of Information Studies, Syracuse University, Syracuse, NY, USA · School of Information, University of Texas at Austin, Austin, TX, USA

  25. Unilateral Relationship Revision Power in Human-AI Companion Interaction

    Mar 24, 2026Benjamin LangeHuman-AI InteractionAI Companions

    Ludwig-Maximilians-Universität München Munich Center for Machine Learning

  26. Technically Love: The Evolution of Human-AI Romance Discourse on Reddit

    Mar 9, 2026Tyler Chang, Jina Huh-Yoo, Afsaneh RaziAI CompanionsHuman-AI Interaction

    Drexel University, Philadelphia, Pennsylvania, USA · Stevens Institute of Technology, Hoboken, New Jersey, USA

  27. Proact-VL: A Proactive VideoLLM for Real-Time AI Companions

    Mar 3, 2026Weicai Yan, Yuhong Dai, Qi Ran +6AI CompanionsVision-Language Agents

    Zhejiang University · Shenzhen University · South China University of Technology +2

  28. How Do Users Negotiate Harmful Value Conflicts with AI Companions? A Study with Minion, a Technology Probe for In-Situ Human-AI Conflict Response

    Nov 11, 2024Qing Xiao, Xianzhe Fan, Xuhui Zhou +4AI CompanionsHuman-AI Interaction

    Human-Computer Interaction Institute, Carnegie Mellon University Pittsburgh, Pennsylvania, USA · The University of Hong Kong Hong Kong SAR, China · Language Technologies Institute, Carnegie Mellon University Pittsburgh, Pennsylvania, USA +2